Polypeptide production at industrial scale currently provides many biologically active polypeptides for a variety of uses, including diagnostic and therapeutic pharmaceuticals, industrial catalysts and nourishment. Polypeptides are produced in a variety of host systems, including transgenic animals and plants, microbes, and cultured mammalian cells. In most cases, the host system is modified by recombinant DNA techniques, for instance resulting in introduction into the host cell of a transgene which encodes a polypeptide of interest. Such a transgene typically includes elements that influence the transcription, translation, and/or processing of the transgene's polypeptide coding sequence. A recombinant host is then identified and isolated which has a suitable yield of a polypeptide of interest, and the cell population of this recombinant host is increased to an extent that it can produce the required amount of polypeptide.
The choice of the host system depends on a number of factors including: (1) the nature and intended use of a polypeptide, and (2) the cost of production. For production of biopharmaceuticals, e.g., therapeutic proteins such as hormones, cytokines, and antibodies, the host system of choice is usually cultured mammalian cells. Considerations with respect to product use and production cost with host cells will be discussed below.
(1) For in vivo therapeutic use, a therapeutic protein must not only have the correct biological activity to alter the course of a disease. It must also do no harm. Most therapeutic proteins are exported from the cell by the secretory pathway. Secreted proteins are modified by a series of post-translational events, including glycosylation, disulfide bond formation, and proteolytic processing. The post-translational modification systems vary among different species and cell types in their detailed mechanisms of action. As a result, the same polypeptide chain can be detectably different when it is produced in different host cells. These differences can be analytical, such as differences in physical properties such as molecular mass, net electrical charge, carbohydrate composition, or structure. The differences can also be functional, affecting for instance the biological activity of the protein itself (catalytic activity, ligand binding characteristics, etc.), and/or its in vivo properties (immunogenicity, biological half life, biodistribution, etc.). Functional differences can, therefore, affect both function and possible side effect(s) of a therapeutic protein. Host cell lines that produce proteins with low efficacy are not suitable for commercial exploitation. Furthermore, host cells which produce modified protein that involves significant side effects in a patient should not be used. These factors are becoming increasingly important considerations during selection of a host cell line for production of a therapeutic protein.
(2) Therapeutic protein production in host cells is an intrinsically costly process. Current methods for industrial production of such proteins often perform poorly, resulting in products that are prohibitively expensive. Poor performance can be due to limitations of protein expression systems and host cell lines currently in use. These limitations mostly have a few specific causes, including (a) failure to identify and isolate recombinant host cell lines that have suitable productivity of proteins (poor predictability), (b) silencing, during the industrial production cycle, of the transgenes that encode proteins (poor stability), and (c) low or incorrect post-translational processing and secretory capacity of the host cell line. These limitations will be considered separately below.
(a) Conventional methods furnish only low frequencies of recombinant host cells that have suitable yields of proteins. Identifying and isolating these rare recombinant cell lines is a laborious and expensive process. The poor predictability of conventional methods means that often a recombinant host cell line is selected for production that has sub-optimal productivity characteristics, simply because a superior recombinant cell line was not identified and isolated during the selection process.
(b) Transgenes are often subject to silencing during cultivation of recombinant host cells. Silencing acts by suppressing transcription of a transgene. Detailed mechanisms of silencing are still not known, and different conventional methods are prone to different kinds of silencing phenomena. With one phenomenon, an individual transgene is silenced by formation of transcriptionally refractory heterochromatin at the transgenic locus. Heterochromatin formation is influenced by the position of genomic integration of a transgene (“position effects” (Boivin and Dura, 1998)). Transgene integration occurs more or less at random. Since most of the genome is heterochromatin, most transgene loci are prone to silencing due to position effects.
A second transgene-silencing phenomenon can occur when two or more copies of a transgene are integrated into a genome during construction of a recombinant cell line. Formation of tandem transgene repeats often occurs during the initial integration step. Furthermore, in order to increase product yield, many recombinant host cell lines are engineered after the integration step to amplify the copy number of a transgene, which also results in tandem transgene repeats (Kaufman, 1990). Tandem repeats and other configurations of multiple transgene copies are particularly prone to silencing (“repeat-induced gene silencing” (Garrick et al., 1998)).
In case that a genome contains multiple copies of a transgene, the yield can also decline via another phenomenon than transcriptional silencing. The number of copies of the transgene can decline during cultivation of a recombinant host cell line. The productivity of such cell lines at the time of selection for use is correlated with a transgene copy number, and consequently as copies of a transgene are lost, the yield declines (Kaufman, 1990).
(c) Different cell types in a mammalian organism have different capacities for post-translational modification and secretion of proteins. The functions of some cell types include production of large quantities of secreted proteins; examples include lymphocytes (producing immunoglobulins), hepatocytes (producing serum proteins), and fibroblasts (producing extracellular matrix proteins). These cell types are favorable sources for deriving host cell lines for production of secreted heterologous proteins. More favorable is the use of a cell line whose progenitor organismal cell type secretes a protein or class of proteins of interest. For example, it is particularly favorable to express recombinant monoclonal antibodies in lymphocytes (or host cells derived from lymphocytes), erythropoietin in hepatocytes (or host cells derived from hepatocytes), and blood clotting factors (e.g., Factor VIII and van Willebrand's factor) in endothelial cells (or host cells derived from endothelial cells).
The use of specific cell types (or cell lines derived therefrom) for production of their affiliated proteins is favorable because such specific cell types will carry out proper post-translational modifications of produced proteins. However, specific cell types often do not have high secretory capacities. For example, cells of the central nervous system, such as neurons, have low intrinsic protein secretion capacities. These cells do secrete proteins, however, including neurotrophins. Neurotrophins regulate the fate and shape of neuronal cells during fetal and juvenile development. Moreover, they influence patterns of neuronal degeneration and regeneration in adults (Bibel and Barde, 2000). Production of neurotrophins for therapeutic applications has considerable biopharmaceutical value (e.g., Axokine™, recombinant ciliary neurotrophic factor from Regeneron). In order to produce heterologous neurotrophins with post-translational modifications (and hence functional properties) that match the naturally-occurring proteins, expression in host cells derived from the central nervous system is favorable. However, production of polypeptides such as neurotrophins in host cell lines such as those derived from neural tissue is inefficient using conventional methods. The predictability of identifying high-expressor isolates from these types of cell lines is often poor; the yield of proteins from such cell lines is generally low, and production levels are characteristically unstable.
Another drawback to a use of specific host cells for production of affiliated proteins is that it is usually difficult to isolate cell lines with favorable biotechnological characteristics. These characteristics for instance include the mode and rate of growth, and the ease of introduction of a transgene. Consequently, various general host cell lines have been established. Examples of these include CHO cells from Chinese hamster ovary (ATCC (American Type Culture Collection) CCL-61), BHK cells from baby hamster kidney (ATCC CCL-10), and Vero cells from African green monkey kidney (ATCC CCL-81). These “general purpose” host cell lines are widely used for production of a number of heterologous proteins. A disadvantage of general purpose cell lines is that the post-translational modifications of heterologous proteins produced by them often differ from the post-translational modifications of the naturally occurring proteins. These differences can have functional consequences resulting in side effects, as discussed above.
Table 1 lists a number of proteins that are currently in use or under development for biopharmaceutical applications. It also lists the tissue or cell type in which each protein is normally produced in the human body. These 24 proteins (or protein classes) come from a wide range of cells and tissue, ranging from highly secretory cells (hepatocytes, endothelial cells) to cells with low secretory capacity (e.g., neural tissue). Currently, neither general-purpose host cells nor specific host cells have qualities that enable optimal expression of the broad spectrum of biopharmaceutically important secreted proteins.
Hence, protein production by conventional host cell lines involves a lot of disadvantages and complications, for instance with respect to yield and post-translational modifications. There is a need in the art for improved protein production in recombinant host cell lines.